Transcript 3rdPlenary_RDA - Research Data Alliance

A Reference Model for
RDA & Global Data Science
Yin Chen
Cardiff University
[email protected]
Wouter Los
University of Amsterdam
[email protected]
1
What is Reference Model?
A standard for description/characterisation of
data, computation of Research Infrastructures
An abstract conceptual Model
captures common requirements
captures state-of-the-art design experiences
With a view of informing future implementation
An ontological framework
A taxonomy of terms, concepts and definitions
A Reference Model for Global data access & sharing
of scientific data
2
Why we need it?
To help the community reach a common vision
To provide a common language for communication
To provide a uniform framework into which RIs’
components can be placed/compared
To provide common solutions to common problems
To secure interoperability
To enable reuse, share of resource/experiences,
avoid duplication efforts
To capture expertise knowledge, state-of-the-art
experience, policies, visions, wisdoms of RDA
To be used as a basis for education of data scientists3
How Shall we Build it?
•
Start from the identification of common
requirements & data lifecycle
Subsystems with points of references between them
4
Common Subsystems
Acquisition -- brings the measures/data streams into
the system (non-reproducible data)
Curation -- manages/maintains quality data
(reproducible data)
Access -- facilities discovery, access (published
data)
Processing -- facilities analysis/mining/experiments
(combined/derived data)
Community Support -- supports users to conduct their
roles in communities (user generated data)
5
Acquisition
Curation
Community
Support
Access
Processing
6
How Shall we build it?
Using Open Distributed Processing (ODP)(ISO/IEC 10746)
A framework for structuring design specification
for large-scale complex distributed systems
An object modelling approach
A viewpoints-based approach to architecture
7
ODP Viewpoints
4/18-20/12
Adapted from ISO/IEC 19793,
2009
Project number: 283465
8